|
{ |
|
"best_metric": 36.38151425762045, |
|
"best_model_checkpoint": "./logs/whisper-tiny-cantonese/checkpoint-800", |
|
"epoch": 4.0275, |
|
"eval_steps": 400, |
|
"global_step": 3200, |
|
"is_hyper_param_search": false, |
|
"is_local_process_zero": true, |
|
"is_world_process_zero": true, |
|
"log_history": [ |
|
{ |
|
"epoch": 0.0, |
|
"learning_rate": 2.1000000000000002e-06, |
|
"loss": 3.4863, |
|
"step": 25 |
|
}, |
|
{ |
|
"epoch": 0.0, |
|
"learning_rate": 4.6e-06, |
|
"loss": 1.6932, |
|
"step": 50 |
|
}, |
|
{ |
|
"epoch": 0.0, |
|
"learning_rate": 7.1e-06, |
|
"loss": 0.6899, |
|
"step": 75 |
|
}, |
|
{ |
|
"epoch": 0.01, |
|
"learning_rate": 9.600000000000001e-06, |
|
"loss": 0.5504, |
|
"step": 100 |
|
}, |
|
{ |
|
"epoch": 0.01, |
|
"learning_rate": 1.2100000000000001e-05, |
|
"loss": 0.5011, |
|
"step": 125 |
|
}, |
|
{ |
|
"epoch": 0.01, |
|
"learning_rate": 1.4599999999999999e-05, |
|
"loss": 0.4671, |
|
"step": 150 |
|
}, |
|
{ |
|
"epoch": 0.01, |
|
"learning_rate": 1.7100000000000002e-05, |
|
"loss": 0.4292, |
|
"step": 175 |
|
}, |
|
{ |
|
"epoch": 0.01, |
|
"learning_rate": 1.9600000000000002e-05, |
|
"loss": 0.4067, |
|
"step": 200 |
|
}, |
|
{ |
|
"epoch": 0.01, |
|
"learning_rate": 2.2100000000000002e-05, |
|
"loss": 0.4019, |
|
"step": 225 |
|
}, |
|
{ |
|
"epoch": 0.01, |
|
"learning_rate": 2.46e-05, |
|
"loss": 0.3764, |
|
"step": 250 |
|
}, |
|
{ |
|
"epoch": 0.01, |
|
"learning_rate": 2.7100000000000005e-05, |
|
"loss": 0.3694, |
|
"step": 275 |
|
}, |
|
{ |
|
"epoch": 0.01, |
|
"learning_rate": 2.96e-05, |
|
"loss": 0.3494, |
|
"step": 300 |
|
}, |
|
{ |
|
"epoch": 0.02, |
|
"learning_rate": 3.21e-05, |
|
"loss": 0.3469, |
|
"step": 325 |
|
}, |
|
{ |
|
"epoch": 0.02, |
|
"learning_rate": 3.46e-05, |
|
"loss": 0.3267, |
|
"step": 350 |
|
}, |
|
{ |
|
"epoch": 0.02, |
|
"learning_rate": 3.71e-05, |
|
"loss": 0.3252, |
|
"step": 375 |
|
}, |
|
{ |
|
"epoch": 0.02, |
|
"learning_rate": 3.960000000000001e-05, |
|
"loss": 0.3167, |
|
"step": 400 |
|
}, |
|
{ |
|
"epoch": 0.02, |
|
"eval_cer": 54.97452400107268, |
|
"eval_loss": 1.3237754106521606, |
|
"eval_runtime": 122.2991, |
|
"eval_samples_per_second": 8.177, |
|
"eval_steps_per_second": 0.515, |
|
"step": 400 |
|
}, |
|
{ |
|
"epoch": 0.02, |
|
"learning_rate": 4.21e-05, |
|
"loss": 0.3117, |
|
"step": 425 |
|
}, |
|
{ |
|
"epoch": 0.02, |
|
"learning_rate": 4.46e-05, |
|
"loss": 0.3074, |
|
"step": 450 |
|
}, |
|
{ |
|
"epoch": 0.02, |
|
"learning_rate": 4.71e-05, |
|
"loss": 0.3039, |
|
"step": 475 |
|
}, |
|
{ |
|
"epoch": 0.03, |
|
"learning_rate": 4.96e-05, |
|
"loss": 0.299, |
|
"step": 500 |
|
}, |
|
{ |
|
"epoch": 0.03, |
|
"learning_rate": 4.994615384615385e-05, |
|
"loss": 0.2914, |
|
"step": 525 |
|
}, |
|
{ |
|
"epoch": 0.03, |
|
"learning_rate": 4.9882051282051286e-05, |
|
"loss": 0.2879, |
|
"step": 550 |
|
}, |
|
{ |
|
"epoch": 0.03, |
|
"learning_rate": 4.981794871794872e-05, |
|
"loss": 0.816, |
|
"step": 575 |
|
}, |
|
{ |
|
"epoch": 0.03, |
|
"learning_rate": 4.975384615384616e-05, |
|
"loss": 0.7056, |
|
"step": 600 |
|
}, |
|
{ |
|
"epoch": 0.03, |
|
"learning_rate": 4.968974358974359e-05, |
|
"loss": 0.6676, |
|
"step": 625 |
|
}, |
|
{ |
|
"epoch": 0.03, |
|
"learning_rate": 4.962564102564103e-05, |
|
"loss": 0.5539, |
|
"step": 650 |
|
}, |
|
{ |
|
"epoch": 1.0, |
|
"learning_rate": 4.956153846153846e-05, |
|
"loss": 0.4248, |
|
"step": 675 |
|
}, |
|
{ |
|
"epoch": 1.0, |
|
"learning_rate": 4.94974358974359e-05, |
|
"loss": 0.2882, |
|
"step": 700 |
|
}, |
|
{ |
|
"epoch": 1.0, |
|
"learning_rate": 4.943333333333334e-05, |
|
"loss": 0.2742, |
|
"step": 725 |
|
}, |
|
{ |
|
"epoch": 1.0, |
|
"learning_rate": 4.936923076923077e-05, |
|
"loss": 0.2565, |
|
"step": 750 |
|
}, |
|
{ |
|
"epoch": 1.01, |
|
"learning_rate": 4.930512820512821e-05, |
|
"loss": 0.2519, |
|
"step": 775 |
|
}, |
|
{ |
|
"epoch": 1.01, |
|
"learning_rate": 4.9241025641025644e-05, |
|
"loss": 0.2437, |
|
"step": 800 |
|
}, |
|
{ |
|
"epoch": 1.01, |
|
"eval_cer": 36.38151425762045, |
|
"eval_loss": 0.7720280289649963, |
|
"eval_runtime": 115.0245, |
|
"eval_samples_per_second": 8.694, |
|
"eval_steps_per_second": 0.548, |
|
"step": 800 |
|
}, |
|
{ |
|
"epoch": 1.01, |
|
"learning_rate": 4.917692307692308e-05, |
|
"loss": 0.2377, |
|
"step": 825 |
|
}, |
|
{ |
|
"epoch": 1.01, |
|
"learning_rate": 4.9112820512820516e-05, |
|
"loss": 0.2217, |
|
"step": 850 |
|
}, |
|
{ |
|
"epoch": 1.01, |
|
"learning_rate": 4.9048717948717955e-05, |
|
"loss": 0.2249, |
|
"step": 875 |
|
}, |
|
{ |
|
"epoch": 1.01, |
|
"learning_rate": 4.898461538461539e-05, |
|
"loss": 0.2165, |
|
"step": 900 |
|
}, |
|
{ |
|
"epoch": 1.01, |
|
"learning_rate": 4.892051282051282e-05, |
|
"loss": 0.212, |
|
"step": 925 |
|
}, |
|
{ |
|
"epoch": 1.01, |
|
"learning_rate": 4.885641025641026e-05, |
|
"loss": 0.2059, |
|
"step": 950 |
|
}, |
|
{ |
|
"epoch": 1.02, |
|
"learning_rate": 4.879230769230769e-05, |
|
"loss": 0.2045, |
|
"step": 975 |
|
}, |
|
{ |
|
"epoch": 1.02, |
|
"learning_rate": 4.872820512820513e-05, |
|
"loss": 0.1901, |
|
"step": 1000 |
|
}, |
|
{ |
|
"epoch": 1.02, |
|
"learning_rate": 4.866410256410256e-05, |
|
"loss": 0.188, |
|
"step": 1025 |
|
}, |
|
{ |
|
"epoch": 1.02, |
|
"learning_rate": 4.86e-05, |
|
"loss": 0.1846, |
|
"step": 1050 |
|
}, |
|
{ |
|
"epoch": 1.02, |
|
"learning_rate": 4.8535897435897435e-05, |
|
"loss": 0.1818, |
|
"step": 1075 |
|
}, |
|
{ |
|
"epoch": 1.02, |
|
"learning_rate": 4.8471794871794874e-05, |
|
"loss": 0.1743, |
|
"step": 1100 |
|
}, |
|
{ |
|
"epoch": 1.02, |
|
"learning_rate": 4.8407692307692306e-05, |
|
"loss": 0.1712, |
|
"step": 1125 |
|
}, |
|
{ |
|
"epoch": 1.02, |
|
"learning_rate": 4.8343589743589746e-05, |
|
"loss": 0.1755, |
|
"step": 1150 |
|
}, |
|
{ |
|
"epoch": 1.03, |
|
"learning_rate": 4.8279487179487185e-05, |
|
"loss": 0.169, |
|
"step": 1175 |
|
}, |
|
{ |
|
"epoch": 1.03, |
|
"learning_rate": 4.821538461538462e-05, |
|
"loss": 0.1602, |
|
"step": 1200 |
|
}, |
|
{ |
|
"epoch": 1.03, |
|
"eval_cer": 43.219808706534366, |
|
"eval_loss": 0.902040421962738, |
|
"eval_runtime": 120.2163, |
|
"eval_samples_per_second": 8.318, |
|
"eval_steps_per_second": 0.524, |
|
"step": 1200 |
|
}, |
|
{ |
|
"epoch": 1.03, |
|
"learning_rate": 4.8151282051282056e-05, |
|
"loss": 0.3491, |
|
"step": 1225 |
|
}, |
|
{ |
|
"epoch": 1.03, |
|
"learning_rate": 4.808717948717949e-05, |
|
"loss": 0.5227, |
|
"step": 1250 |
|
}, |
|
{ |
|
"epoch": 1.03, |
|
"learning_rate": 4.802307692307693e-05, |
|
"loss": 0.5003, |
|
"step": 1275 |
|
}, |
|
{ |
|
"epoch": 1.03, |
|
"learning_rate": 4.795897435897436e-05, |
|
"loss": 0.4226, |
|
"step": 1300 |
|
}, |
|
{ |
|
"epoch": 1.03, |
|
"learning_rate": 4.78948717948718e-05, |
|
"loss": 0.3688, |
|
"step": 1325 |
|
}, |
|
{ |
|
"epoch": 2.0, |
|
"learning_rate": 4.783076923076923e-05, |
|
"loss": 0.1807, |
|
"step": 1350 |
|
}, |
|
{ |
|
"epoch": 2.0, |
|
"learning_rate": 4.776666666666667e-05, |
|
"loss": 0.1726, |
|
"step": 1375 |
|
}, |
|
{ |
|
"epoch": 2.0, |
|
"learning_rate": 4.7702564102564104e-05, |
|
"loss": 0.1553, |
|
"step": 1400 |
|
}, |
|
{ |
|
"epoch": 2.0, |
|
"learning_rate": 4.763846153846154e-05, |
|
"loss": 0.1564, |
|
"step": 1425 |
|
}, |
|
{ |
|
"epoch": 2.01, |
|
"learning_rate": 4.757435897435898e-05, |
|
"loss": 0.1497, |
|
"step": 1450 |
|
}, |
|
{ |
|
"epoch": 2.01, |
|
"learning_rate": 4.7510256410256415e-05, |
|
"loss": 0.1424, |
|
"step": 1475 |
|
}, |
|
{ |
|
"epoch": 2.01, |
|
"learning_rate": 4.744615384615385e-05, |
|
"loss": 0.1374, |
|
"step": 1500 |
|
}, |
|
{ |
|
"epoch": 2.01, |
|
"learning_rate": 4.738205128205128e-05, |
|
"loss": 0.137, |
|
"step": 1525 |
|
}, |
|
{ |
|
"epoch": 2.01, |
|
"learning_rate": 4.731794871794872e-05, |
|
"loss": 0.1396, |
|
"step": 1550 |
|
}, |
|
{ |
|
"epoch": 2.01, |
|
"learning_rate": 4.725384615384615e-05, |
|
"loss": 0.1326, |
|
"step": 1575 |
|
}, |
|
{ |
|
"epoch": 2.01, |
|
"learning_rate": 4.718974358974359e-05, |
|
"loss": 0.1352, |
|
"step": 1600 |
|
}, |
|
{ |
|
"epoch": 2.01, |
|
"eval_cer": 52.596764101188874, |
|
"eval_loss": 0.7722536325454712, |
|
"eval_runtime": 116.9997, |
|
"eval_samples_per_second": 8.547, |
|
"eval_steps_per_second": 0.538, |
|
"step": 1600 |
|
}, |
|
{ |
|
"epoch": 2.02, |
|
"learning_rate": 4.712564102564103e-05, |
|
"loss": 0.1236, |
|
"step": 1625 |
|
}, |
|
{ |
|
"epoch": 2.02, |
|
"learning_rate": 4.706153846153846e-05, |
|
"loss": 0.1264, |
|
"step": 1650 |
|
}, |
|
{ |
|
"epoch": 2.02, |
|
"learning_rate": 4.69974358974359e-05, |
|
"loss": 0.116, |
|
"step": 1675 |
|
}, |
|
{ |
|
"epoch": 2.02, |
|
"learning_rate": 4.6933333333333333e-05, |
|
"loss": 0.1207, |
|
"step": 1700 |
|
}, |
|
{ |
|
"epoch": 2.02, |
|
"learning_rate": 4.686923076923077e-05, |
|
"loss": 0.1173, |
|
"step": 1725 |
|
}, |
|
{ |
|
"epoch": 2.02, |
|
"learning_rate": 4.6805128205128205e-05, |
|
"loss": 0.1123, |
|
"step": 1750 |
|
}, |
|
{ |
|
"epoch": 2.02, |
|
"learning_rate": 4.6741025641025644e-05, |
|
"loss": 0.1109, |
|
"step": 1775 |
|
}, |
|
{ |
|
"epoch": 2.02, |
|
"learning_rate": 4.667692307692308e-05, |
|
"loss": 0.1111, |
|
"step": 1800 |
|
}, |
|
{ |
|
"epoch": 2.02, |
|
"learning_rate": 4.6612820512820516e-05, |
|
"loss": 0.1111, |
|
"step": 1825 |
|
}, |
|
{ |
|
"epoch": 2.03, |
|
"learning_rate": 4.654871794871795e-05, |
|
"loss": 0.1075, |
|
"step": 1850 |
|
}, |
|
{ |
|
"epoch": 2.03, |
|
"learning_rate": 4.648461538461539e-05, |
|
"loss": 0.1071, |
|
"step": 1875 |
|
}, |
|
{ |
|
"epoch": 2.03, |
|
"learning_rate": 4.642051282051283e-05, |
|
"loss": 0.4093, |
|
"step": 1900 |
|
}, |
|
{ |
|
"epoch": 2.03, |
|
"learning_rate": 4.635641025641026e-05, |
|
"loss": 0.3722, |
|
"step": 1925 |
|
}, |
|
{ |
|
"epoch": 2.03, |
|
"learning_rate": 4.62923076923077e-05, |
|
"loss": 0.3687, |
|
"step": 1950 |
|
}, |
|
{ |
|
"epoch": 2.03, |
|
"learning_rate": 4.622820512820513e-05, |
|
"loss": 0.2899, |
|
"step": 1975 |
|
}, |
|
{ |
|
"epoch": 3.0, |
|
"learning_rate": 4.616410256410257e-05, |
|
"loss": 0.1995, |
|
"step": 2000 |
|
}, |
|
{ |
|
"epoch": 3.0, |
|
"eval_cer": 28.69401984446232, |
|
"eval_loss": 0.5907071828842163, |
|
"eval_runtime": 121.9491, |
|
"eval_samples_per_second": 8.2, |
|
"eval_steps_per_second": 0.517, |
|
"step": 2000 |
|
}, |
|
{ |
|
"epoch": 3.0, |
|
"learning_rate": 4.61e-05, |
|
"loss": 0.1061, |
|
"step": 2025 |
|
}, |
|
{ |
|
"epoch": 3.0, |
|
"learning_rate": 4.603589743589744e-05, |
|
"loss": 0.103, |
|
"step": 2050 |
|
}, |
|
{ |
|
"epoch": 3.0, |
|
"learning_rate": 4.5971794871794874e-05, |
|
"loss": 0.0999, |
|
"step": 2075 |
|
}, |
|
{ |
|
"epoch": 3.01, |
|
"learning_rate": 4.590769230769231e-05, |
|
"loss": 0.0993, |
|
"step": 2100 |
|
}, |
|
{ |
|
"epoch": 3.01, |
|
"learning_rate": 4.5843589743589746e-05, |
|
"loss": 0.0931, |
|
"step": 2125 |
|
}, |
|
{ |
|
"epoch": 3.01, |
|
"learning_rate": 4.577948717948718e-05, |
|
"loss": 0.0922, |
|
"step": 2150 |
|
}, |
|
{ |
|
"epoch": 3.01, |
|
"learning_rate": 4.571538461538462e-05, |
|
"loss": 0.0872, |
|
"step": 2175 |
|
}, |
|
{ |
|
"epoch": 3.01, |
|
"learning_rate": 4.565128205128205e-05, |
|
"loss": 0.0923, |
|
"step": 2200 |
|
}, |
|
{ |
|
"epoch": 3.01, |
|
"learning_rate": 4.558717948717949e-05, |
|
"loss": 0.0874, |
|
"step": 2225 |
|
}, |
|
{ |
|
"epoch": 3.01, |
|
"learning_rate": 4.552307692307692e-05, |
|
"loss": 0.0892, |
|
"step": 2250 |
|
}, |
|
{ |
|
"epoch": 3.01, |
|
"learning_rate": 4.545897435897436e-05, |
|
"loss": 0.0845, |
|
"step": 2275 |
|
}, |
|
{ |
|
"epoch": 3.02, |
|
"learning_rate": 4.539487179487179e-05, |
|
"loss": 0.0856, |
|
"step": 2300 |
|
}, |
|
{ |
|
"epoch": 3.02, |
|
"learning_rate": 4.533076923076923e-05, |
|
"loss": 0.0785, |
|
"step": 2325 |
|
}, |
|
{ |
|
"epoch": 3.02, |
|
"learning_rate": 4.526666666666667e-05, |
|
"loss": 0.0785, |
|
"step": 2350 |
|
}, |
|
{ |
|
"epoch": 3.02, |
|
"learning_rate": 4.5202564102564104e-05, |
|
"loss": 0.0779, |
|
"step": 2375 |
|
}, |
|
{ |
|
"epoch": 3.02, |
|
"learning_rate": 4.513846153846154e-05, |
|
"loss": 0.0758, |
|
"step": 2400 |
|
}, |
|
{ |
|
"epoch": 3.02, |
|
"eval_cer": 39.57271833378028, |
|
"eval_loss": 0.7932679057121277, |
|
"eval_runtime": 122.4295, |
|
"eval_samples_per_second": 8.168, |
|
"eval_steps_per_second": 0.515, |
|
"step": 2400 |
|
}, |
|
{ |
|
"epoch": 3.02, |
|
"learning_rate": 4.5074358974358976e-05, |
|
"loss": 0.0735, |
|
"step": 2425 |
|
}, |
|
{ |
|
"epoch": 3.02, |
|
"learning_rate": 4.5010256410256415e-05, |
|
"loss": 0.0709, |
|
"step": 2450 |
|
}, |
|
{ |
|
"epoch": 3.02, |
|
"learning_rate": 4.494615384615385e-05, |
|
"loss": 0.0752, |
|
"step": 2475 |
|
}, |
|
{ |
|
"epoch": 3.03, |
|
"learning_rate": 4.4882051282051286e-05, |
|
"loss": 0.0723, |
|
"step": 2500 |
|
}, |
|
{ |
|
"epoch": 3.03, |
|
"learning_rate": 4.481794871794872e-05, |
|
"loss": 0.0673, |
|
"step": 2525 |
|
}, |
|
{ |
|
"epoch": 3.03, |
|
"learning_rate": 4.475384615384616e-05, |
|
"loss": 0.2012, |
|
"step": 2550 |
|
}, |
|
{ |
|
"epoch": 3.03, |
|
"learning_rate": 4.468974358974359e-05, |
|
"loss": 0.3141, |
|
"step": 2575 |
|
}, |
|
{ |
|
"epoch": 3.03, |
|
"learning_rate": 4.462564102564103e-05, |
|
"loss": 0.3052, |
|
"step": 2600 |
|
}, |
|
{ |
|
"epoch": 3.03, |
|
"learning_rate": 4.456153846153847e-05, |
|
"loss": 0.243, |
|
"step": 2625 |
|
}, |
|
{ |
|
"epoch": 3.03, |
|
"learning_rate": 4.44974358974359e-05, |
|
"loss": 0.203, |
|
"step": 2650 |
|
}, |
|
{ |
|
"epoch": 4.0, |
|
"learning_rate": 4.443333333333334e-05, |
|
"loss": 0.0757, |
|
"step": 2675 |
|
}, |
|
{ |
|
"epoch": 4.0, |
|
"learning_rate": 4.436923076923077e-05, |
|
"loss": 0.0709, |
|
"step": 2700 |
|
}, |
|
{ |
|
"epoch": 4.0, |
|
"learning_rate": 4.4305128205128205e-05, |
|
"loss": 0.0623, |
|
"step": 2725 |
|
}, |
|
{ |
|
"epoch": 4.0, |
|
"learning_rate": 4.424102564102564e-05, |
|
"loss": 0.0662, |
|
"step": 2750 |
|
}, |
|
{ |
|
"epoch": 4.01, |
|
"learning_rate": 4.417692307692308e-05, |
|
"loss": 0.0617, |
|
"step": 2775 |
|
}, |
|
{ |
|
"epoch": 4.01, |
|
"learning_rate": 4.4112820512820516e-05, |
|
"loss": 0.0584, |
|
"step": 2800 |
|
}, |
|
{ |
|
"epoch": 4.01, |
|
"eval_cer": 31.465093411996065, |
|
"eval_loss": 0.7228736281394958, |
|
"eval_runtime": 121.4415, |
|
"eval_samples_per_second": 8.234, |
|
"eval_steps_per_second": 0.519, |
|
"step": 2800 |
|
}, |
|
{ |
|
"epoch": 4.01, |
|
"learning_rate": 4.404871794871795e-05, |
|
"loss": 0.0548, |
|
"step": 2825 |
|
}, |
|
{ |
|
"epoch": 4.01, |
|
"learning_rate": 4.398461538461539e-05, |
|
"loss": 0.0579, |
|
"step": 2850 |
|
}, |
|
{ |
|
"epoch": 4.01, |
|
"learning_rate": 4.392051282051282e-05, |
|
"loss": 0.0623, |
|
"step": 2875 |
|
}, |
|
{ |
|
"epoch": 4.01, |
|
"learning_rate": 4.385641025641026e-05, |
|
"loss": 0.0581, |
|
"step": 2900 |
|
}, |
|
{ |
|
"epoch": 4.01, |
|
"learning_rate": 4.379230769230769e-05, |
|
"loss": 0.0588, |
|
"step": 2925 |
|
}, |
|
{ |
|
"epoch": 4.01, |
|
"learning_rate": 4.372820512820513e-05, |
|
"loss": 0.053, |
|
"step": 2950 |
|
}, |
|
{ |
|
"epoch": 4.02, |
|
"learning_rate": 4.3664102564102563e-05, |
|
"loss": 0.0574, |
|
"step": 2975 |
|
}, |
|
{ |
|
"epoch": 4.02, |
|
"learning_rate": 4.36e-05, |
|
"loss": 0.0517, |
|
"step": 3000 |
|
}, |
|
{ |
|
"epoch": 4.02, |
|
"learning_rate": 4.3535897435897435e-05, |
|
"loss": 0.0536, |
|
"step": 3025 |
|
}, |
|
{ |
|
"epoch": 4.02, |
|
"learning_rate": 4.3471794871794874e-05, |
|
"loss": 0.0492, |
|
"step": 3050 |
|
}, |
|
{ |
|
"epoch": 4.02, |
|
"learning_rate": 4.3407692307692314e-05, |
|
"loss": 0.0498, |
|
"step": 3075 |
|
}, |
|
{ |
|
"epoch": 4.02, |
|
"learning_rate": 4.3343589743589746e-05, |
|
"loss": 0.044, |
|
"step": 3100 |
|
}, |
|
{ |
|
"epoch": 4.02, |
|
"learning_rate": 4.3279487179487185e-05, |
|
"loss": 0.0486, |
|
"step": 3125 |
|
}, |
|
{ |
|
"epoch": 4.03, |
|
"learning_rate": 4.321538461538462e-05, |
|
"loss": 0.0505, |
|
"step": 3150 |
|
}, |
|
{ |
|
"epoch": 4.03, |
|
"learning_rate": 4.315128205128206e-05, |
|
"loss": 0.0477, |
|
"step": 3175 |
|
}, |
|
{ |
|
"epoch": 4.03, |
|
"learning_rate": 4.308717948717949e-05, |
|
"loss": 0.0485, |
|
"step": 3200 |
|
}, |
|
{ |
|
"epoch": 4.03, |
|
"eval_cer": 40.93143827657102, |
|
"eval_loss": 0.7707632780075073, |
|
"eval_runtime": 123.6857, |
|
"eval_samples_per_second": 8.085, |
|
"eval_steps_per_second": 0.509, |
|
"step": 3200 |
|
} |
|
], |
|
"logging_steps": 25, |
|
"max_steps": 20000, |
|
"num_train_epochs": 9223372036854775807, |
|
"save_steps": 800, |
|
"total_flos": 1.008359991853056e+19, |
|
"trial_name": null, |
|
"trial_params": null |
|
} |
|
|